Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    punctuation and disagreement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     চক্ষে
    0.41
    綃
    0.41
     прадстаў
    0.37
    ESTAMP
    0.36
    튿
    0.36
     dechlor
    0.35
    Ꮡ
    0.35
     mantle
    0.35
    ецца
    0.35
    攒
    0.34
    POSITIVE LOGITS
    our
    0.41
    ai
    0.38
     refused
    0.38
     terrorists
    0.38
     Idris
    0.37
    ars
    0.37
     disagreed
    0.36
    int
    0.36
     di
    0.36
     politicians
    0.35
    Act Density 0.001%

    No Known Activations